gpu: nvgpu: move resetup_ramfc to common channel

On gp10b, ramfc contains information related to syncpoint
protection, which restricts the syncpoint increment operation
to a safe set of syncpoints. This information must be
updated when a syncpoint is assigned to a channel.

Added the following ramfc HALs
- ramfc.get_syncpt
- ramfc.set_syncpt

And replaced
- fifo.resetup_ramfc

With
- channel.set_syncpt

Use new ramfc HALs, move resetup_ramfc implementation
from fifo to common channel code:
- nvgpu_channel_set_syncpt

NVGPU-1750

Change-Id: I036a0b7b2d9fd6ccd9f30094ae33e6c38a96e0cc
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2075938
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Thomas Fleury
2019-03-18 12:51:18 -07:00
committed by mobile promotions
parent 1701a267bc
commit 6009662fa5
16 changed files with 86 additions and 61 deletions

View File

@@ -1029,8 +1029,8 @@ static int nvgpu_ioctl_channel_get_user_syncpoint(struct channel_gk20a *ch,
}
nvgpu_mutex_release(&ch->sync_lock);
if (g->ops.fifo.resetup_ramfc) {
err = g->ops.fifo.resetup_ramfc(ch);
if (g->ops.channel.set_syncpt) {
err = g->ops.channel.set_syncpt(ch);
if (err)
return err;
}